JP2000029517A - Travel control device for autonomous vehicles - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reproduce a taught traveling route while securing stability in steering without increasing a track error caused by the error of a sensor for detecting a its own position. SOLUTION: At the time of play back, car body control is performed as if its own vehicle 1 is hauled from a virtual traction point VTP and even when the position of the virtual traction point VTP is dispersed by the detection error of the present position, while reducing the dispersion on the side of the present vehicle 1 caused by the length of a traction distance, namely, lever ratio, stability in steering is secured without increasing the track error.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an autonomous traveling vehicle having a teaching mode in which the vehicle travels while storing the self-positions of a plurality of points and the vehicle body orientation, and a playback mode in which the traveling locus of the teaching mode is reproduced. The present invention relates to a travel control device.

[0002]

2. Description of the Related Art In recent years, autonomous vehicles have been developed that can perform mowing and lawn mowing work in various fields such as golf courses, river banks, parks, and the like unattended. Such an autonomous traveling vehicle often repeats a fixed operation such as lawn mowing work in the same area. Therefore, the operation traveling route is stored in advance, and the stored operation traveling route is reproduced to perform the autonomous operation. Running, so-called teaching (teaching)
There are many systems that employ a playback (playback) method of travel control.

For example, JP-A-59-105112 discloses that each target point has a distance from a preceding target point, a reference azimuth which is an angle between a geomagnetic direction and a traveling direction, and a reference azimuth which is an angle between the traveling direction and the next target point. A technique is disclosed in which a traveling azimuth angle, which is an angle with a direction, is preliminarily taught, and during playback traveling, data that has been taught is read, and when a deviation from the read data occurs, the vehicle travels while correcting the deviation. ing.

Japanese Patent Laid-Open Publication No. Hei 3-135606 discloses a teaching method for setting a reference direction of two directions, forward and backward, in a reciprocating operation of a field work vehicle by manually traveling an arbitrary one stroke or a plurality of strokes for an arbitrary time. Discloses a technique to be performed.

[0005]

However, since the sensor for detecting the self-position includes an error, at the time of teaching, as shown in FIG. The stored data will be stored.

For this reason, if the deviation between the self-position detected by the sensor and the self-position taught during the playback is reduced to reproduce the traveling route faithfully during playback, the steering gain must be increased, and the error of the sensor must be increased. As a result, the steering system reacts to cause blurring, and the running trajectory meanders, thereby lowering stability.

The present invention has been made in view of the above circumstances, and is intended to reproduce a taught traveling route while ensuring steering stability without increasing a trajectory error due to an error of a sensor for detecting a self-position. It is an object of the present invention to provide a travel control device for an autonomous vehicle that can be used.

[0008]

According to the first aspect of the present invention,
In a traveling control device of an autonomous traveling vehicle having a teaching mode in which the vehicle travels while storing its own position and the body direction of a plurality of points, and a playback mode in which a traveling locus of the teaching mode is reproduced, in the playback mode, A virtual tow point separated by a predetermined distance from each of the self positions stored in the teaching mode is set, and steering and speed control equivalent to when the vehicle is towed from each virtual tow point is performed to perform the teaching mode. It is characterized by comprising means for reproducing the running locus.

According to a second aspect of the present invention, in the first aspect of the present invention, the virtual tow point is set based on a self-position of each point stored in the teaching mode and a body direction. .

According to a third aspect of the present invention, in the first or second aspect of the present invention, the vehicle body target direction is calculated from the virtual tow point and the current self-position, and the vehicle body direction of the own vehicle is set to the vehicle body target direction. It is characterized in that steering control is performed so as to be in the azimuth.

[0011] The invention according to claim 4 is the invention according to claims 1, 2, 3
In the invention described in any one of the above, speed control is performed such that a distance between the virtual tow point and the host vehicle is within a set range.

That is, according to the first aspect of the present invention, a virtual towing point which is separated from the self-position of the plurality of points stored in the teaching mode by a predetermined distance from the self-position in the playback mode is set. When the vehicle is towed from the virtual tow point, steering and speed control equivalent to the towing are performed to reproduce the taught traveling locus.

In this case, according to the second aspect of the present invention, the virtual tow point is set based on the self-position of each point and the vehicle body direction stored in the teaching mode. The vehicle body target direction is calculated from the point and the current own position, and the steering control is performed so that the vehicle direction of the own vehicle becomes the vehicle body target direction. Further, in the invention according to claim 4, speed control is performed so that the distance between the virtual tow point and the host vehicle is within the set range.

[0014]

Embodiments of the present invention will be described below with reference to the drawings. 1 to 10 relate to an embodiment of the present invention, FIG. 1 is an explanatory diagram showing reproduction of a traveling locus by a virtual traction point, and FIG. 2 is a lawn mowing vehicle equipped with a D-GPS mobile station and a D-GPS. FIG. 3 is a block diagram of a control device, FIG. 4 is a flowchart showing a control routine in a teaching mode, FIGS. 5 and 6 are flowcharts showing a control routine in a playback mode, FIG. Is an explanatory diagram of a virtual tow point, FIG. 8 is an explanatory diagram showing a virtual tow point for each playback pointer, FIG. 9 is an explanatory diagram showing an area where the virtual tow point is located, and FIG. FIG.

In FIG. 2 (a), reference numeral 1 denotes an unmanned and self-propelled autonomous traveling work vehicle. In this embodiment, a mowing blade 2 such as a mower provided at the lower part of the vehicle body is used to cut grass and grass on a golf course. This is a lawn mowing vehicle that performs lawn mowing work. The lawn mowing vehicle 1 transmits the driving force of the engine to the driving wheels via a hydrostatic continuously variable transmission (HST) using a swash plate type piston pump / motor, and runs by rear wheel steering. .

The lawn mowing vehicle 1 also includes various devices related to self-position detection necessary for autonomous traveling, that is, a satellite radio receiver for receiving a radio wave from a satellite and measuring its own position, and a traveling history. A sensor for dead reckoning navigation for measuring the current position based on the current position, a sensor for detecting a traveling obstacle, and the like are mounted.

As a satellite radio receiver, a global positioning system (hereinafter referred to as GPS) is used.
(Abbreviated as "abbreviated") to receive a radio wave from a GPS satellite and measure its own position.
This GPS receiver performs position observation at a fixed station located at a known point and feeds back correction information (differential information) to the mobile station, that is, movement for so-called differential GPS (hereinafter abbreviated as D-GPS). It is a station GPS receiver.

For this reason, the mowing vehicle 1 includes the mobile station GP.
An antenna 3 of the S receiver and an antenna 4 of a wireless communication device for receiving differential information from a fixed station are provided upright, and at a known point outside the vehicle, as shown in FIG. A fixed station 30 having an antenna 31 of a fixed station GPS receiver and an antenna 32 of a wireless communication device for transmitting differential information to the mobile station GPS receiver is arranged.

The dead-reckoning sensors include a two-axis geomagnetic sensor 5 as an azimuth detecting sensor and a wheel encoder 6.
Is installed to correct the output of the geomagnetic sensor 5,
An inclination sensor 7 for detecting a roll angle and a pitch angle of the vehicle body is installed substantially at the center of the vehicle body. As sensors for detecting obstacles, non-contact sensors 8a and 8b such as ultrasonic sensors and optical sensors are attached to the front and rear portions of the lawn mowing vehicle 1, and contact sensors 9a and 9b using micro switches and the like. Are attached to the front and rear ends of the lawn mower 1.

The lawn mowing vehicle 1 described above is controlled by a control device 50 composed of a microcomputer or the like. As shown in FIG. 3, the control device 50 includes a dead reckoning position detecting unit 51 that performs positioning based on signals from the geomagnetic sensor 5 and the wheel encoder 6, and a D-position that performs positioning based on information from a GPS satellite group. GPS position detector 52, non-contact sensors 8a and 8b, and contact sensors 9a and 9b
Detection unit 5 that detects obstacles based on signals from the
3. a work data storage unit 54 for storing various work data and maps, a travel control unit 55 for performing travel control using data from the detection units 51, 52, 53 and data from the work data storage unit 54, a travel control A vehicle control unit 56 for controlling the vehicle in accordance with an instruction from the unit 55, a drive control unit 57 for driving each mechanism unit based on an output from the vehicle control unit 56, a steering control unit 58, a cutting blade control unit 59. And so on.

The dead reckoning position detector 51 calculates the traveling distance by integrating the vehicle speed detected by the wheel encoder 6, and associates the traveling distance with a change in the traveling direction detected based on the output of the geomagnetic sensor 5. By accumulating, the traveling history from the reference point is calculated, the current position of the vehicle 1 is measured, and the positioning data is output to the traveling control unit 55. The azimuth detection error caused by the biaxial terrestrial magnetism sensor 5 detecting components other than the horizontal component of the terrestrial magnetism vector due to the road surface gradient is corrected by the roll angle and the pitch angle of the vehicle body by the inclination sensor 7. Further, the sensor connected to the dead reckoning position detector 51 is not limited to the combination of the geomagnetic sensor 5 and the wheel encoder 6, but may be a gyro or the like.

In the D-GPS position detecting section 52, the mobile station G
Navigation messages from GPS satellites (at least four for three-dimensional positioning, at least three for two-dimensional positioning) captured via a PS receiver, ie, satellite clock correction coefficients, orbit information, The position of the vehicle 1 is measured from the positioning information such as the satellite calendar and the satellite arrangement, and the differential information from the fixed station 30 received via the wireless communication device, and the positioning data is output to the travel control unit 55. .

The fixed station 30 for the D-GPS position detecting section 52 includes a D-GPS fixed station section 33 having a fixed station GPS receiver, and D-GPS information for transmitting the differential information from the D-GPS fixed station section 33. Transmission section 3
4 and so on. The D-GPS fixed station section 34 processes positioning information from the GPS satellites received via the fixed station GPS receiver to create differential correction data. This differential correction data is
The data is converted into wireless communication packet data in the D-GPS information transmitting unit 34 and transmitted to the D-GPS position detecting unit 52 of the lawn mowing vehicle 1.

In this embodiment, the fixed station 30 of the D-GPS is installed as a specific device intended for the mobile station of the lawnmower 1. However, the radio station for transmitting differential information is used. It is also possible to use an existing D-GPS fixed station equipped with the above, or an existing D-GPS fixed station that transmits differential information via a communication satellite.

In the obstacle detecting section 53, the non-contact type sensor 8
a, 8b and the contact type sensors 9a, 9b detect unpredictable obstacles, and output a detection signal to the travel control unit 55.

The work data storage unit 54 stores fixed data such as terrain data of a work area where grass and lawn mowing work is performed, terrain data of the whole area including a plurality of work areas, data obtained by processing signals from each sensor, D- Positioning data by GPS, positioning data by dead reckoning, correction data of D-GPS / dead reckoning, own vehicle 1 calculated based on this correction data
The present position data, data at the time of teaching traveling described later, and the like are stored.

On the other hand, the traveling control unit 55 detects the positioning data from the D-GPS position detecting unit 52 and the dead reckoning position detection in order to cope with a decrease in the positioning accuracy of the D-GPS due to the state of capturing satellites and the state of reception of radio waves. The difference from the positioning data from the unit 51 is accumulated and averaged, and the self-position is calculated by correcting the positioning data by the dead reckoning position detection unit 52 with the averaged value. Then, an error amount between the current self-position and the target position is calculated with reference to the data in the work data storage unit 54, and a vehicle control instruction to the vehicle control unit 56 is determined. still,
When an obstacle is detected by the obstacle detection unit 54,
The vehicle control unit 56 is instructed to stop the vehicle.

In this embodiment, when the self-position (X, Y) of the lawn mowing vehicle 1 is measured, the absolute origin (0, 0) of the X, Y coordinate axes is set to the position of the D-GPS fixed station 30, and Y The coordinate axis is taken in the north direction, the X coordinate axis is taken in the east direction, and the azimuth of the vehicle body is measured in a clockwise direction with reference to the north.

The vehicle control unit 56 converts the instruction from the traveling control unit 55 into a specific control instruction amount, and
7. Output to the steering control unit 58 and the cutting blade control unit 59. Accordingly, the drive control unit 57 controls the hydraulic motor 10 to generate hydraulic pressure to be supplied to each unit, and controls the drive mechanism units 11 such as a throttle actuator, a brake actuator, and an HST hydraulic control actuator for controlling engine output. Drive. The steering control unit 58 drives the rear wheel steering mechanism unit 13 via the rear wheel steering hydraulic control valve 12 to perform steering control, and the cutting blade control unit 59 controls the cutting blade control hydraulic control valve 14. Servo control of the cutting blade mechanism 15 is performed via this.

Here, the control device 50 includes a mode (teaching mode) for storing a travel route during manual operation,
There is a mode (playback mode) in which the stored traveling route is reproduced to perform autonomous traveling. When the lawn mowing vehicle 1 is switched to a teaching mode by an operation panel (not shown) and a teaching number is set, self-moving at a plurality of points in the teaching traveling is performed. Position, body direction, direction of travel (forward, backward)
And the like, are sampled at predetermined time intervals, and cutter operation events such as raising and lowering and rotating the cutter 2 that occur during traveling are additionally recorded.

Then, the data collection by the teaching driving is completed, and thereafter, the mode is switched to the playback mode.
When a work program instruction (playback instruction) is executed by designating a teaching number, playback traveling is autonomously executed based on the teaching data of the designated number, and the traveled trajectory is reproduced.

Hereinafter, the control processing in the teaching mode and the control processing in the playback mode by the control device 50 will be described with reference to flowcharts shown in FIGS.

First, in the control routine in the teaching mode shown in FIG. 4, in step S101, necessary initial parameters are set by the teaching number designated when the teaching mode is entered. The initial parameters are a sampling interval ttime of traveling data in units of a predetermined time (for example, 10 msec), an absolute position (Xini, Yini) of the own vehicle 1 at the start of teaching, and the like, and are added to the beginning of teaching data. . Then, following this initial parameter, data sampled by the following processing is added and stored as packet data for each teaching number. The running data sampling interval ttime is a unique machine parameter (M / C parameter) that differs for each model.

Then, the process proceeds to step S102, where the D-GPS
The running data such as the current own position (Xorg, Yorg) of the host vehicle 1 measured by combining the above and dead reckoning, the body direction Θ detected based on the output of the geomagnetic sensor 5, and the running direction (forward / backward) are stored in the memory. To memorize. In addition, own vehicle 1
Is actually stored as an offset from the absolute position (Xini, Yini) at the start of teaching to save memory, but in the following description, the self-position (Xorg, Yorg) is The description will be made as an absolute position from (D-GPS fixed station 30).

The forward / reverse running direction is determined by the state of the pedal (HST pedal) connected to the input shaft of the HST. That is, in the HST mounted on the lawn mowing vehicle 1 of the present embodiment, the swash plate angle of the HST is proportional to the vehicle speed, and the forward and backward movement can be switched by the direction of the supplied hydraulic pressure. It is possible to determine whether the vehicle is moving forward or backward by checking whether the position is the position (automatic braking state), the forward position, or the backward position.

In the following step S103, the state of a cutter operation switch (not shown) is read, and whether or not a cutter operation event has occurred, and the event type and time of occurrence of the event are stored in a memory. The cutter operation event includes raising and lowering, forward rotation, reverse rotation, and stop of the cutter 2, and the event occurrence time is recorded retroactively to the time when the switch is turned on.

Next, the process proceeds to step S104, where it is determined whether or not the driver has terminated the teaching mode based on the state of a switch on an operation panel (not shown). If the teaching mode has not been terminated, the process proceeds to step S105 to perform sampling. It is checked whether or not the next traveling data sampling time determined by the interval ttime has been reached.

As a result, when the sampling time of the next traveling data has not reached, it is checked in step S106 whether or not the recording time of the next cutter operation event has been reached. Returning to step S105, when the next cutter operation event recording time has been reached, the process returns to step S103 to record a change in event content or occurrence of a new event.

Then, proceeding to step S104, when the teaching mode is not ended, it is checked in step S105 whether or not the sampling time of the next traveling data has been reached. When the sampling time of the next traveling data has been reached, Returning to step S102, the next traveling data is sampled.

That is, since the cutter operation event occurs arbitrarily, recording is performed at a time period shorter than the sampling interval of the traveling data. For example, if the time unit of the sampling interval ttime is 10 msec and t
When time = 100, a cutter operation event is recorded every 10 msec, and 10 msec × 100 = 1 sec
The running data is recorded every time.

Then, the above processing is repeated to sample the traveling data and record the cutter operation event. When the teaching mode ends in step S104,
An end mark is added to the end of the sampled teaching data to end the routine, and exit from the teaching mode.

Next, the processing at the time of playback will be described. As shown in FIG. 7, the traveling locus reproduced during playback does not use the position recognition origin of the lawn mowing vehicle 1 as a control target, but moves the lawn mowing vehicle 1 to the traveling direction side (when the vehicle 1 moves forward, A virtual traction point VTP separated by a set distance (virtual traction distance) length is set forward and backward of the host vehicle 1 and the virtual traction point VTP is reproduced as a control target. Position of virtual tow point (Xvtp, Y
vtp) uses the self-position (Xorg, Yorg) sampled in the teaching mode and the body direction Θ,
It is calculated by the following equations (1) and (2). Xvtp = Xorg + length · sinΘ (1) Yvtp = Yorg + length · cosΘ (2)

At the time of playback, as shown in FIG. 1, the vehicle body control is performed as if the own vehicle 1 is being towed from the virtual towing point VTP, and the self-position at the time of playback is the self-position at the time of teaching. If it is out of the range, the self-position during playback and the virtual traction point VT
The steering control is performed in the direction of the straight line connecting P and the traveling locus during teaching is reproduced.

That is, at the time of playback, as shown in FIG. 8, for each time pointer (playback pointer) t in the control routine, the body position data P (t) and the body direction A (t) sampled in the teaching mode are set. , The virtual tow point VTP (t) is calculated one after another, and by controlling the steering amount and speed with respect to the virtual tow point VTP (t), control equivalent to movement between two points where the target position constantly changes is performed. In addition, the steering stability is ensured without increasing the trajectory error.

Hereinafter, the control routine of FIGS. 5 and 6 based on the playback command will be described. In this routine, first, in step S201, initial parameters necessary for playback traveling are set. This initial parameter is
Control time interval pbtime during playback, distance teng from teaching data to virtual tow point VTP
th, the distance to the in-position area that is the target position range during playback (im-position distance during playback) pb
length, the size pbinpos of the in-position area during playback, the size pbstop of the deceleration area for performing deceleration control during playback, and the HST for speed control.
One step amount pbacc for increasing or decreasing the pedal angle is an M / C parameter specific to the model.

In order to execute the playback instruction, the teaching number and the maximum depression amount of the HST pedal are required as operands. By specifying the maximum depression amount of the HST pedal, regardless of the contents of the teaching data,
The speed increase value during speed control is limited.

Next, the process proceeds to step S202, and refers to the time pointer (playback pointer) t in the control routine, and corresponds to the current playback pointer t from the teaching data stored at each sampling interval ttime. Search for position data in teaching data. In this case, the sampling interval t during teaching
From the time, control interval pbtime during playback
Is short and there is no data directly corresponding to the current playback pointer t, linear interpolation is performed from the preceding and succeeding teaching data, and the linearly interpolated data is used as teaching data corresponding to the current playback pointer t.

In the following step S203, the self-position (X) in the teaching data corresponding to the playback pointer t
org, Yorg), the vehicle azimuth Θ, and the model-specific virtual traction distance “length”, and the position (Xvtp, Y) of the virtual traction point VTP (t) at the playback pointer t.
vtp) is calculated according to the above-described equations (1) and (2).

Then, the process proceeds to step S204, where the D-GPS
Own position (X, Y) of the own vehicle 1 and position of virtual tow point (Xvtp, Yv)
tp) is calculated as the vehicle body target direction, and in step S205, the steering amount is determined from the vehicle body target direction and the current vehicle body direction. Then, a steering command is output to the steering control unit 58 to servo-control the rear wheel steering mechanism unit 13 via the rear wheel steering hydraulic control valve 12 to perform steering of the rear wheel that is a steering wheel.

Thereafter, the process proceeds to step S206, where the virtual tow point VTP (t) calculated from the teaching data based on the current self-position data is stored in the area A, as shown in FIG.
It is determined which of B, C, and D is located, speed control is performed according to the determination result, and the progress of the playback pointer t is adjusted.

That is, the playback pointer t for reproducing the teaching data is basically made to advance in the real time at which the teaching is performed. However, the control of the HST pedal and the playback pointer t Coordination of progress is required. For this reason, the areas A, B, C, and D are divided based on the vehicle body position at the time of playback.

Each of the areas A, B, C, and D is based on a position separated from the current position recognition origin of the own vehicle 1 by a playback in-position distance pblenth ahead in the traveling direction.
The area of pbinpos in front of this reference position is
A region (C region; imposition region) that is within an allowable range with respect to the virtual towing distance tlength, which is the target position, a region farther than the inposition region is a D region, and a region having a width pbstop closer to the imposition region. Is a B region (deceleration region), and a region closer to the B region is an A region.

When the virtual tow point VTP (t) calculated from the data at the time of teaching is in the area A, the flow proceeds to step S207, and when it is in the area B, the flow proceeds to step S2.
Go to 09. When the virtual tow point VTP (t) is in the region C, the process proceeds to step S211. When the virtual tow point VTP (t) is in the region D, the process proceeds to step S213.

When the virtual traction point VTP (t) is in the area A, the current position of the host vehicle 1 is changed to the virtual traction point VTP.
It is determined that the vehicle is approaching (t) too much and the position at the time of teaching has been excessively far, and in step S207, the speed of the own vehicle 1 is set to 0 and the own vehicle 1 is forcibly stopped. Then, the process proceeds to step S208 to advance the control time pointer (playback pointer) t by the set value.
Proceed to step S215. Normally, the playback pointer t advances by the control time interval pbtime,
In this case, the playback pointer t is advanced by one.

When the virtual traction point VTP (t) is in the area B, the current position of the host vehicle 1 is changed to the virtual traction point VTP.
Although it is close to (t), it is determined that the position has exceeded the position at the time of teaching, and the HST pedal angle is operated in step S209 to set the host vehicle 1 to the set rate (1 of the increase or decrease of the HST pedal angle).
(Step amount) Decrease by pbacc. However, the minimum value of the speed due to the deceleration is 0, and the speed is not reversed (the vehicle does not move backward). Then, in step S210, the control time pointer (playback pointer) t is advanced by the set value, and the flow advances to step S215.

If the virtual tow point VTP (t) is in the area C, it is determined that the current position of the host vehicle 1 is substantially at the time of teaching, and the current vehicle speed is maintained in step S211.
In step S212, the control time pointer (playback pointer) t is advanced by the set value, and the flow advances to step S215.

On the other hand, when the virtual tow point VTP (t) is in the area D, the current position of the host vehicle 1 is changed to the virtual tow point VT.
It is determined that the position is far from P (t) and has not reached the position at the time of teaching, and the pedal angle of the HST is operated to increase the speed of the host vehicle 1 by the set rate pbacc in step S213 (however, the playback command The specified HST pedal angle does not exceed). Then, in step S214, the control time pointer (playback pointer) t is held without advancing, and the process proceeds to step S215.

In step S215, the cutter operation event is reproduced while performing the playback traveling control. The cutter operation event may be reproduced at a control interval specified by a control time interval pbtime during playback with respect to a recording time unit of the event, and the cutter operation is controlled for each event type. At this time, the signals from the non-contact sensors 8a and 8b and the signals from the contact sensors 9a and 9b are checked, and when an obstacle is detected, the rotation of the cutter 2 is immediately stopped and the vehicle is stopped. Then, if an obstacle is detected even after the lapse of a predetermined time, a warning to notify an abnormality is issued and measures such as stopping the engine are taken.

Thereafter, the flow advances to step S216 to check whether or not the teaching data reaches the end mark and the data ends. As a result, if the teaching data is not completed, it is checked in step S217 whether or not the next control time has come. If the next control time has not come, step S2
When a time waiting loop is reached at 17 and the next control time is reached, the process returns to step S202 to continue the above processing.

Then, the reproduction of the traveling locus and the cutter operation event at the time of teaching is continued. When the teaching data reaches the end mark in step S216, it is determined that the teaching data has ended, the routine is ended, and the playback mode is exited.

FIG. 10 shows an example of reproduction of the teaching locus, for performing steering and speed control equivalent to when the lawn mowing vehicle 1 is towed by the virtual tow point VTP which is separated from the lawn mowing vehicle 1 by a predetermined distance. In contrast, the detection error of the self-position and the body direction is amplified and does not appear on the reproduction trajectory as in the related art.
Even if the position of the TP varies, the variation on the lawn mowing vehicle 1 side can be reduced and the steering stability can be improved due to the length of the towing distance, that is, the lever ratio.

[0062]

As described above, according to the present invention, a virtual tow point which is separated by a predetermined distance from each self position in the playback mode is set for each self position of a plurality of points stored in the teaching mode. Then, when the vehicle is towed from each virtual tow point, the steering and speed control equivalent to the towing is performed to reproduce the taught trajectory, so that the influence of the detection error of the own position is reduced by the length of the towing distance. It is possible to obtain excellent effects such as the ability to reduce the trajectory and to reproduce the teaching trajectory while ensuring the steering stability without increasing the trajectory error.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram showing reproduction of a traveling locus by a virtual tow point.

FIG. 2 is a lawn mower equipped with a D-GPS mobile station and D
-Explanatory drawing which shows the fixed station for GPS

FIG. 3 is a block diagram of a control device.

FIG. 4 is a flowchart showing a control routine in a teaching mode.

FIG. 5 is a flowchart showing a control routine in a playback mode.

FIG. 6 is a flowchart showing a control routine in a playback mode (continued).

FIG. 7 is an explanatory diagram of a virtual traction point.

FIG. 8 is an explanatory diagram showing virtual tow points for each playback pointer.

FIG. 9 is an explanatory diagram showing an area where a virtual tow point is located;

FIG. 10 is an explanatory diagram showing a traveling locus during playback traveling.

FIG. 11 is an explanatory diagram showing a traveling locus in a conventional playback traveling.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Lawn mowing vehicle 51 ... Dead reckoning position detection part 52 ... D-GPS position detection part 55 ... Travel control part VTP ... Virtual traction point length ... Virtual traction distance

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 2B043 AA04 AB07 AB15 BA02 BA09 BB11 EA13 EA14 EA37 EB04 EB05 EB08 EB09 EB15 EB17 EC12 EC13 ED12 2B083 AA02 BA18 5H301 AA01 AA03 AA09 BB12 CC03 CC06 DD02GG06 GG06 GG06 HH04 JJ03

Claims (4)

[Claims] 1. A travel control device for an autonomous traveling vehicle having a teaching mode in which the vehicle travels while storing the self-positions of the plurality of points and a vehicle body direction, and a playback mode in which a travel locus of the teaching mode is reproduced. In the playback mode, virtual towing points separated by a predetermined distance from each of the self positions stored in the teaching mode are set, and steering and speed control equivalent to when the vehicle is towed from each of the virtual towing points is performed. A travel control device for reproducing the travel locus in the teaching mode. 2. The travel control device for an autonomous vehicle according to claim 1, wherein the virtual tow point is set based on a self-position and a body direction of each point stored in the teaching mode. 3. The vehicle body target direction is calculated from the virtual tow point and the current own position, and steering control is performed so that the vehicle direction of the own vehicle becomes the vehicle body target direction. Item 3. A travel control device for an autonomous vehicle according to Item 2. 4. The travel control of an autonomous vehicle according to claim 1, wherein speed control is performed such that a distance between the virtual tow point and the own vehicle is within a set range. apparatus.